RF Jamming Attacks and Countermeasures in Wireless Vehicular Networks
Dimitrios Kosmanos, Antonios Argyriou · 2021
This chapter introduces a cross-layer intrusion detection system (IDS) that is based on supervised learning algorithms k-Nearest Neighbors, random forests, and a data fusion method that combines the outcomes of the two classification algorithms. Specifically, it proposes the use of radio frequency (RF) signals for estimating the relative speed between the jammer and the receiver in the physical layer. The chapter investigates how the proposed variations of relative speed (VRS) metric between the jammer and the receiver can enhance the detection accuracy of different types of RF jamming attacks. It evaluates the proposed IDS under different “smart” jamming strategies and a different number of interfering nodes in the area. Many countermeasures have been proposed to address the RF jamming issues from different technical perspectives. Using the proposed VRS metric among the other features for training and testing we achieved an increase of RF jamming detection accuracy over 10% under certain conditions and “smart” jamming strategies.